OpenAI's new voice model brings GPT-5-level reasoning to real-time conversations

OpenAI launched GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper for live voice applications.

The models add stronger reasoning, better tool use, and low-latency streaming transcription for spoken conversations.

GPT-Realtime-Translate supports translation across 70+ languages, expanding real-time multilingual voice use cases.

The release brings voice AI closer to text AI, improving support, translation, and workflow automation tasks.